Application of the Month
Backlash Measurement in Worm Reducers
May, 2014 - View all Application Examples
Being heavily involved in the integration of gearboxes in motion control applications, DieQua is often approached with requests for solutions that meet various backlash requirements. Sometimes these requirements are quite stringent, which we have products for. Other times a high degree of accuracy is not needed but a reasonable limit is necessary.
The second scenario is typical when customers are interested in controlling costs while still getting acceptable repeatability and positioning. This is when we suggest some our hybrid solutions that include a gearbox more typically used for standard power transmission applications and an adapter for accepting a servo motor.
Especially for lighter duty applications, our Varvel series of worm reducers is a common consideration. It is relatively lower in cost than other gear technologies, has a compact right angle configuration, is shaft mountable and has a unique input coupling and flange system that lends itself to accommodating a variety of motors.
That all being said, it is still subject to some of the inherent problems standard worm designs have in controlling backlash, although not to the level of some competitive products. The issues are of consistent component tolerances and axial worm play.
Let’s face it, most industrial worm reducers are low cost, volume production products. While today’s automated machines are faster and more accurate, there is still a range of tolerances for worm, ring and housing dimensions that can result in a sloppy gearbox. And with more of them being produced in a certain country known for high volume production, the quality from several manufacturers isn’t exactly improving.
The typical procedure for measuring backlash in most gearbox types is to lock the input and measure the play with an indicator on a lever arm mounted to the output shaft. Moving the lever arm up and down until the gear clearance is removed, and without instilling torsion on gear teeth or bearings, will provide a linear value that can be translated into an approximate angular value.
So, why is measuring backlash in worm boxes a different animal than most other gearbox types? It’s the construction and support of the worm shaft that alters the measurement procedure.
Most gearboxes have either loaded angular contact bearings, or in the case of helical products, radial bearings with close tolerance races in precise bearing bores. This resists separation forces which keeps the gears where they are supposed to be, allowing gear clearance to be the only element affecting backlash.
Worm boxes function a little differently. Considering that the worm shaft acts like a screw there is a linear reaction force during operation in addition to the radial force. That is where the potential problem lies.
The position of worm shaft support bearings is usually maintained with snap rings. And for a snap ring to be installed there has to be some clearance between the bearing race and the snap ring groove. This clearance allows a certain amount of worm shaft axial play that directly adds to lost motion in reversing applications.
So, after that long winded introduction, how to measure backlash in a worm reducer? It takes at least 2 indicators, or even better 3.
The different scenario affecting worm reducers is that you can’t lock the input shaft like for other gearbox types. You still measure on the output side but you also have to measure movement on the input shaft.
As you can see in the included graphic, along with the output indicator there is an indicator on the end of the worm shaft. If, when moving the output lever arm, there is also an axial movement on worm shaft, this situation is adding to the overall backlash level.
While with most manufacturers you get what you get, Varvel offers additional hand fitting of shims to eliminate as much of the axial play as possible. This drastically can reduce the measured backlash values by up to half, achieving levels of 12-20 arc minutes depending on size. This is a significant improvement not available from other manufacturers of similar products.
So, I mentioned that a third indicator might be necessary. Where would that go? Well, that would also go on the input shaft. While worm boxes resist back driving, there is a possibility of some input rotary movement when testing the output lever arm. Therefore, in addition to measuring input axial play, its important to assure no input rotary motion as this would skew the measurement results.
A final point to consider is the output shaft itself. Most worm reducers have a hollow output shaft as standard and use insert shafts when necessary. These are typically keyed and are captured with a retaining kit. Be sure the insert shaft is tightly retained so you aren’t adding key play to the output measurement.
Yes, I know this seems a lot of work to measure the backlash on an inexpensive gearbox. But if the proper performance can be extracted, a lot of money can be saved over purchasing other types of gearboxes.
Director of Marketing
Did you know that DieQua publishes a monthly email newsletter that features interesting articles just like this one?
Our newsletter was designed to be a little different than other manufacturer's newsletters that are primarily product centric. While products and services available from DieQua are highlighted, other elements revolve around the personal and professional development of the design engineer.
To receive this newsletter, please visit the registration page and fill out your information.